Advances in Tunneled Central Venous Cathetersfor Dialysis: Design and Performance

Stephen R. AshClarian Arnett Health and Wellbound, Inc., Ash Access Technology and HemoCleanse, Inc., Lafayette, andPurdue University, West Lafayette, Indiana

ABSTRACT

Over 70% of patients initiating chronic hemodialysis in theUnited States have a tunneled central venous catheter (CVC)for dialysis as their first blood access device. Tunneled CVChave requirements that are unparalleled by other accessdevices: high blood flow rates at moderate pressure drops with-out obstruction, minimal trauma to the vein, resistance toocclusion by fibrous sheathing, prevention of infection, avoid-ance of clotting, biocompatibility, avoidance of lumen collapseand kinking and breaks, resistance to antiseptic agents, place-

ment with minimal trauma, and radiopaque appearance onX-ray. This publication reviews the numerous designs fortunneledCVCand evaluates the advantages and disadvantagesof each design. A catheter that self-centers in the superiorvena cava (Centros�) is described, along with early clinicalresults. Current challenges and future directions for tunneledCVC for dialysis are discussed, included means to diminishcatheter-related infections, catheter tip clotting, fibrous sheath-ing, central venous stenosis, and external component bulk.

Prevalence and Problems of Central VenousCatheters for Hemodialysis

Over 70% of patients initiating chronic hemodialysisin theUnited States have a tunneled central venous cath-eter (CVC) as their first blood access device (1). Asshown by Centers for Medicare Services (CMS) and inFig. 1, at 90 days of dialysis the catheter is still the accessof choice, used in patients in whom the fistula or grafthas notmatured, was not workable, or was not indicated(2). Over the period from 2002 to 2005 the number ofgrafts being used at 90 days decreased and the numberof fistulas increased, but the percentage of cathetersbeing used remained about the same.

Many patients would be better served if an arteriove-nous (AV) fistula had been placed some months earlier,and if it were fully developed and functional when dialy-sis was implemented. However much this is the desiredcourse, it is not often the actual course. In all of thesingle-minded enthusiasm of the Fistula First program,we sometimes forget that tunneled CVC are used inpatients starting dialysis because they offer advantages.As summarized by Beathard (3,4) tunneled CVC for

dialysis have advantages (noted in Table 1) as well assome significant disadvantages:

The requirements for a tunneled CVC for dialysis areactually multiple and stringent, unlike requirements forany other access device (5):

• High blood flow rates at moderate pressuredrops, with few instances of outflow failure andpressure alarms regardless of patient fluid statusand catheter position relative to the vein wall.

• Minimal trauma to the vein intima to avoidthrombosis and venous stenosis.

• Resistance to occlusion by fibrous sheathing.• Prevention of bacterial migration around the

catheter after placement.• Avoidance of contamination of the catheter

lumen.

Fig. 1. Distribution of access types 90 days after initiation of

chronic outpatient dialysis (CMS 2006 report [2]).

Address correspondence to: Stephen R. Ash, MD, FACP,3601 Sagamore Parkway North, Suite B, Lafayette, IN 47904,or e-mail: sash@ashaccess.com.

Seminars in Dialysis—2008DOI: 10.1111/j.1525-139X.2008.00494.xª 2008 Copyright the Author.Journal compilation ª 2008 Wiley Periodicals, Inc.

1

REDUCING TUNNELED HEMODIALYSIS CATHETER MORBIDITY

• Avoidance of seeding of the outside of the cathe-ter during bacteremia.

• Avoidance of clotting at the tip or within thecatheter.

• Biocompatibility of the catheter surfaces, avoid-ing removal of white cells or platelets.

• Avoidance of lumen collapse under negativepressure.

• Avoidance of kinking of catheter segments atpoints of bending.

• Physical strength and integrity to avoid breaksor disconnections of any component (ability toreplace broken connectors is desirable).

• Resistance to antiseptic agents that might beapplied at the skin exit site.

• Placement procedures with minimal trauma,difficulty and risk.

• Radiopaque appearance on X-ray, for evalua-tion of location during placement and after use.

Each tunneled CVC has a risk of failing one or moreof these requirements, and each failure results in signifi-cantmedical problems.The use of dual lumen CVC for removing and return-

ing blood during dialysis is commonplace now but in thelate 1970s this concept revolutionized dialysis (6). Beforethe development of CVC for dialysis, dialysis was possi-ble only with an arterial access, through an inter-nal ⁄external AV silicone shunt or through separatecatheters placed into an artery and a vein and removedafter each treatment.The development of CVC for dialysis was not simple,

especially for single-body catheters. Drawing bloodfrom a central vein at 200–400 ml ⁄min is a delicate andsomewhat unpredictable process. The pressure in centralveins is much lower than in arteries and vein walls arethinner and more distensible, even though the flow ofblood through central veins is the same as through cen-tral arteries. Removal of blood through the ports of aCVC in a vein creates a negative pressure around theseports because of direct suction and because of the Ber-

noulli effect. This negative pressure can cause the veinwall to collapse around the ports and obstruct flow intothe ports, even if the flow through the vein is muchhigher than the flow of blood through the catheter. If afibrous tissue sheath forms around the catheter andreaches the tip or if clots form around the tip, the entryport to the catheter becomes smaller and the velocity ofblood flow is increased. The increased blood velocity cre-ates a greater negative pressure around the ports, andincreases the tendency to pull the vein wall over the tip.There are four classic approaches to the problem of

providing sufficient blood outflow through dual-lumenCVC for dialysis:

• Place the removal and return lumens within theright atrium, where the tips cannot rest against avenous wall and only one lumen usually restsagainst the atrial wall.

• Position the catheter with the removal lumen onthe inside of the catheter curve, directing thislumen away from the vein wall and toward thebloodstream.

• Use a large catheter so that the removal lumencannot be blocked by a small clot or a smallamount of fibrous tissue.

• Provide multiple blood entry ports in all direc-tions around the circumference of each cathetertip so that some of the ports are always facingaway from the vein wall.

There are problems of and limitations on each of theseapproaches. Positioning the tips of the removal andreturn lumens at the middle of the atrium is somewhatdifficult, especially as the relative positions of the cathe-ter and the heart change when the patient stands up afterlying on the procedure table, and as the removal lumenis shorter than the return lumen. Positioning the catheterso that the removal lumen is on the inside of the cathetercurve is not always easy as the catheter course throughthe subcutaneous tissue and central veins is rather com-plex and tortuous. Placing a larger catheter is alwaysmore difficult and somewhat more traumatic than plac-ing a smaller catheter, especially if the larger catheter isnot round in shape. Providing multiple side holes in alldirections around the catheter tips requires that twocatheters be placed, or that one catheter must separateinto two separate tips. Side holes in a catheter also havedisadvantages. If they are too large or too many, bloodwill quickly flow through the tip of the catheter afterplacement and between uses, removing catheter locksolutions and promoting clotting at the tip. If the sideholes are too small or too few then bloodwill flow in andout only through the tip of the catheter, thus diminishingany advantage of the side holes. Further, any single-body catheter that becomes covered by a sheath will losefunction, whether there are side holes or not. Sheathingof catheters occurs only where the catheter contacts avein or atrial wall (7).When sheathing develops it is diffi-cult to correct by tissue plasminogen activator infusion,stripping or catheter replacement (8).A newer solution to the problem of obtaining unre-

stricted blood removal through the catheter is to designthe distal part of the catheter so that it rests against thewall of the superior vena cava (SVC) and supports the

TABLE 1. Advantages and disadvantages of tunneled central venous

catheters for dialysis

Advantages Disadvantages

Universally applicable (functionalin nearly 100% of patients)

High morbidity caused bythrombosis and infection

Ability to insert into multiple sites Risks of permanent centralvenous stenosis orocclusion

Maturation time not required Discomfort and cosmeticdisadvantage of externalappliance

Venipuncture not required Lower blood flow rates,requiring longer dialysistimes

No hemodynamic consequences(no CP recirculation)Ease and low cost of placementand replacementAbility to provide access over aperiod of monthsEase of correcting thromboticcomplications

2 Ash

removal lumen in the center of the vena cava. Thisapproach, used in the Centros� catheter (Angio-dynamics, Inc., Queensbury, NY), is described below.

Types of CVC for Dialysis and a Short History

Central venous catheter for dialysis are classified intoeither ‘‘acute’’ or ‘‘chronic’’ catheters, depending onwhether the catheters are expected to be used for onlyseveral days or months to years. Acute CVC aredesigned to be placed with a minimum amount of effort.Historically, acute catheters for dialysis were relativelyrigid, pointed catheters with a conically shaped tip thatcould be advanced into the vein directly over a guide-wire. The catheter body dilates the entry site as it isadvanced into the vein. Acute CVC for dialysis have nosubcutaneous cuff or locking device and a short lineartunnel.More recently some acute catheters have becomeavailable with soft tips similar to that of chronic tun-neled catheters. The tract around the guidewire is dilatedand the catheter is stiffened using a stylet, then advancedover a guidewire in a manner similar to over-the-wireplacement of a tunneled catheter (described below).

Tunneled CVC for dialysis are soft, blunt-tippedcatheters with a subcutaneous Dacron� ‘‘cuff’’(INVISTAS.a r.l.,Wichita,Kansas) for tissue in-growthor a plastic ‘‘grommet’’ to immobilize the cathetersbelow the skin surface. Tunneled CVC are generally

placed through internal jugular veins into the SVC withthe goal of placing the tips of the catheter at the junctionof the SVC and the right atrium. Alternative venousaccess points are external jugular veins, subclavian veins,and femoral veins. Due to their blunt shape tunneledCVC have traditionally been placed through a ‘‘splitsheath,’’ which is a cylindrical thin-walled plastic deviceadvanced into the vein over a dilator. The dilator has acentral lumen that follows the guidewire. The guidewireand dilator are then removed and the split sheath open-ing is closed with a finger or valve to prevent excessivebleeding. The catheter is then inserted through the splitsheath into the central vein. The split sheath is split alongtwo preformed grooves, and the halves are retractedaround the catheter, leaving the catheter in positionwithin the central vein. More recently, techniques havebeen developed to allow placement of tunneled CVC tobe performed over a guidewire placed through a previ-ously dilated tract, in a manner similar to acute CVC fordialysis. A plastic catheter or stylet within the catheterstiffens the catheter to allow it to follow the guidewiremore easily.

Tunneled CVC for dialysis have a curved subcutane-ous tunnel leading from the vein insertion site to a dis-tant exit site. The cuff or plastic grommet fixes thecatheter in position and prevents bacteria at the exit sitefrommigrating around the catheter. The cuff also servesas the outer limit for a fibrous tunnel that leads to theentrance point of the central vein. This tunnel is similar

Fig. 2. Various designs of tunneled central venous catheters for dialysis.

ADVANCES IN TUNNELED CVC FOR DIALYSIS 3

to a vein wall and is contiguous with the internal jugularvein (or other vein of insertion), creating a passagewayfor blood or air when the catheter is removed. The tun-nel stops at the Dacron cuff where it melds into thefibrous tissue surrounding the cuff. Without the cuff, asin acute catheters, this tunnel continues all of the way tothe skin exit site over time, creating potential for back-and-forth movement of the catheter and potential peri-catheter bacterial migration around the catheter.A pictorial history of tunneled dialysis catheters is

included in Fig. 2, and this history is discussedmore fullyin a recent review (5). Canaud et al. devised a cathetersystem composed of two 10 French catheters, eachplaced into the vena cava and with tips leading to theright atrium. Flow rate was excellent over many monthsof use (9,10). Tesio added subcutaneous cuffs and thecatheter became more popular. More recent versions ofthe Canaud catheters have included a subcutaneousplastic grommet to fix the catheter limbs in place; theSchon catheter has a similar device. Quinton designedthe PermCath dual lumen chronic catheter, an oval-shaped chronic catheter of about 20 French circumfer-ence and including two cylindrical 8 French lumens(11,12).Mahurkar designed a chronic CVCof softmate-rials and blunt tips and double-d blood flow lumens(13). The Ash Split Cath chronic catheter has a double-d

configuration in the mid-body, but separates into twoseparate distal tips, each with side holes in all directionsThe Palindrome catheter is a double-d catheter withboth lumens having the same length, but with oppositelyangled and symmetrical side ports (14). The Centros�catheter has outward bends of the tips that contact theinferior vena cava in two places and inward bends toplace the arterial and venous ports in the middle of thevena cava. No side holes are needed as the ports do notrest on the venous wall (15).

Advantages and Disadvantages of VariousTunneled CVC Designs

Hydraulic Performance of Tunneled CVCVersus Grafts and Fistulas with Needles

In spite of the wide variety of designs of tunneledCVC for dialysis there are few comparative studies todefine advantages of one design over another. Thebest way to characterize the effectiveness of flow in adialysis access is to determine the ‘‘conductance’’ ofthe access, which is the flow rate divided by the pres-sure drop on the arterial (blood removal) limb (4,5).Merely describing the achieved blood flow rate wasduring an entire dialysis is not very descriptive, as theblood flow rate depends upon many factors such asnumber of pressure alarms, volume status of thepatient, physician’s prescription, pressure gradients,etc. There have been few comparative studies showinghow the hydraulic conductance of catheters comparesto needles, though one by Twardowski in 1999 washelpful. In this publication he showed that althoughthere is considerable scatter, the hydraulic conduc-tance of most catheters is similar to a 16 gauge needlein a graft or fistula during dialysis.

Split-Tip Catheters Versus Single-Body Step-Tips

The basic concept of the Split Cath is to provide sideholes around each limb of the catheter, similar to aCanaud or Tesio catheter. This assures that even if eachlimb lies against the surface of the vena cava or atriumthat some side holes will be facing the lumen, and awayfrom the wall. In vitro studies demonstrate the hydraulicadvantage of this design, using models as shown inFig. 3. Mareels et al. performed studies using computa-tional flow dynamics and particle imaging and demon-strated that the Split Cath design had a considerablylower shear rate than any other catheter design, withonly 32% of the tip portion having a shear stress over10 Pa as shown in Fig. 4. However, the downside of theSplit Cath design was that it also creates areas of stagna-tion in the tip, with blood residence time at this locationover 0.03 seconds (16).Clinically which is better, catheters with a split tip or a

single body? Several studies have shown a slight advan-tage of the Split Cath over step-tip catheters. Trerotolaet al. performed a randomized study of 12 end-stagerenal disease (ESRD) patients receiving 14 F Split Cathcatheters placed versus 12 patients receiving 13.5 FHickman catheters (17). Weekly for 6 weeks the blood

Fig. 2. (Continued)

4 Ash

flow rate was measured using Transonic flow monitors,while the blood pump was set at speeds of 200, 300, 350,400 ml ⁄minute and as high as possible with sustainedflow. The measured blood flow rate at the highest pumpsetting was 422 � 12 ml ⁄minute for the Split Cath and359 � 13 for the Hickman (p < 0.005) as shown inFig. 5. Recirculation was significantly less at all pumpsettings for the Split Cath patients (p = 0.01–0.06),though for both catheters it remained below 6% asshown in Fig. 6.

Long-term functional survival of CVC is probably themost significant measure of the success of their use. Oneproblemwith such studies is a lack of firm definitions forpatency failure or catheter-related bloodstream infection(CRBSI), to serve as endpoints. Many catheters are

removed in these studies for presumed failure to flow orsepsis. In the Trerotola study above, the Split Cath hadslightly better 6 week survival than the Hickman cathe-ter. Richard et al. performed a randomized study com-paring the Split Cath, Opti-Flow, and Tesio catheters in113 placements in ESRDpatients (18).Maximum (effec-tive) blood flow rates were compared between the cathe-ters immediately after placement, and 30 and 90 daysafter placement. Blood flow rate tended to be higherwith the Split Cath but results were not significantly dif-ferent. Failure-free survival of the catheters was ana-lyzed with an average follow-up of 120 days. Thoughstatistically not significant, predicted lifespan appearedhigher for the Split Cath and Tesio catheters than theOpti-Flow. Placement complications occurred only withTesio and Opti-Flow catheters. These results are shownin Fig. 7.

Fig. 3. Types of catheters evaluated in CFD and PIV studies

(16). (A) Cut straight; (B) cut straight with slot; (C) cut angle; (D)

cut straight with hole; (E) concentric; (F) cut angle with hole; (G)

Ash split.

Fig. 4. Measured shear rate and residence time for various parts of various catheter designs (16).

Fig. 5. Relationship between roller pump setting of dialysis

machine and actual flow rate by Transonic� device, for Split

Catheter and Bard Hickman (17).

ADVANCES IN TUNNELED CVC FOR DIALYSIS 5

Trerotola et al. also performed a randomizedstudy comparing the Split Cath and Opti-Flow cath-eters in 132 placements in ESRD patients (19). Com-plications during placement were no different for thetwo catheters and ranged 15–17% (mostly, kinking).Opti-Flow delivered significantly higher flow rateswhen tested at 1 month, but there was no significantdifference in flow at 6 months. Recirculation wasalways lower with the Split Cath catheter but notalways significantly different. The Split Cath had sig-nificantly longer half-life, partly due to lower infec-tion rate but also due to some mechanical failures ofthe Opti-Flow. Postulating on reasons that the SplitCath might have lower infection rates, the authorssuggested that the ‘‘self-cleaning’’ function of theSplit Cath, with high velocity flow through the sideholes created by the ‘‘step down’’ tip, may diminishfibrin sheath and therefore decrease the opportunityfor bacterial colonization. The results of seven non-randomized studies of the Split Cath confirmingcatheter survival rates of about 9 months on averageare summarized in our review in Seminars in Dialysis(5). However, the overall longevity of Split Caths isnot any better than that of Tesio catheters, whichhave been shown to have up to 2 years of assistedfunction (20).

Side Holes Versus No Side Holes in TunneledCVC

As discussed above regarding the Split Cath design,catheter side holes have advantages and disadvantages.They can decrease the overall resistance and providelower shear rate and better flow in the short run, espe-cially on the arterial side. However, they also can createrelative dead space at the end of the tip promoting clot-ting. They allow flow of blood through the tip to washout locking solution almost as soon as the catheter isfilled with it (5,21,22). Side holes also allow clots toadhere to the catheter tip, a process facilitated by lumenswith rough edges. Tesio ⁄Canaud catheters have six sideholes in a spiral shape. It is not uncommon for thesecatheters to clot completely within several hours afterplacement. Removing the clot with forceful irrigation,using the catheter for dialysis and then re-locking thecatheter with heparin usually results in a catheter thatfunctions for a long time. Presumably this is because thecatheter becomes ‘‘biolized’’ or protein-coated and isthereforemore resistant to clotting.Side holes can also increase overall catheter resistance.

To understand this, it is important to understand moreabout how blood flows within a catheter. A curious phe-nomenon is that the relationship between flow and pres-sure is not the same on both lumens of the catheter. Asshown in Fig. 8, the higher the blood flow rate, thehigher the hydraulic resistance on the arterial side. Therelationship between flow and pressure is not a straightline as it is on the venous lumen (23). The best explana-tion for this phenomenon is given by Fricker et al., whodemonstrated that at the arterial inlet port, blood entersthe catheter as a ‘‘developing’’ flow pattern, not a para-bolic or developed flow pattern, as shown in Fig. 9. This‘‘plug’’ flow results in higher shear at the catheter innersurface. The higher the flow rate, the greater the distanceit takes the blood to develop parabolic flow within thecatheter lumen, and the greater the hydraulic resistanceof the catheter (24). Also, the more complicated the tipshape, the greater the distortion of flow and the greaterthe distance of developing flow, as shown in Fig. 10. Ifthe tip is blocked and all of the flow is through the sideholes, then hydraulic resistance can increase markedly

Fig. 8. Relationship between flow and pressure during in vitro

experiments with DD catheters. The hydraulic resistance of the

venous lumen is fairly constant, but that of the arterial lumen

increases with increasing flow (23).

Fig. 6. Recirculation in forward flow measured by Transonic,

for recently placed Split Caths versus Bard Hickman (17).

Fig. 7. Survival of catheters to primary failure, for Split Cath,

Optiflow, and Tesio catheters (19).

6 Ash

with blood flow rate, regardless of the size of the sideholes (24).

Do side holes provide any advantage clinically?Tal et al. performed a prospective study of Mahur-kar-type single-body tunneled catheters, comparingcatheters with side holes to catheters without sideholes (25). On removal, many of the catheters withside holes had adherent clots, while those withoutside holes had fewer adherent clots, as shown inFig. 11. In a follow-up of over 12 weeks there wasa slightly higher flow rate for catheters with sideholes, though without a significant difference. Sur-prisingly there was a significantly higher rate inci-

dence of CRBSI in the catheters with side holesversus those without (2.54 ⁄1000 catheter days versus0.254 ⁄1000 catheter days). The authors surmisedthat clots adherent to the catheter tip served as anidus for infection, after seeding from systemic bac-teremia or lumen contamination.

Thus, side holes have advantages and disadvantages.If a catheter limb rests against a vein or atrial wall andthe tip is blocked by sheath or thrombus, side holes allowcontinued flow though at a higher hydraulic resistance.However if a catheter tip can be positioned away from awall, as in the atrium or within the SVC blood stream,then side holes would not be necessary and in fact disad-vantageous.

Symmetric Tipped Tunneled CVC, thePalindromeTM

In 2005 Tal reported on a new catheter design withsymmetric tips and biased ports, as shown in Fig. 12(14). In an animal study the percentage recirculation wascompared to that of step-tip and split-tip catheters,immediately after placement. All of the catheters wererun in reverse flow and tips placed in the SVC or theright atrium. As shown in Fig. 13, the Palindrome hadless recirculation than any of the other catheters. Sur-prisingly, all catheters had slightly more recirculationwhen placed in the atrium than in the SVC. The step-tip

Fig. 9. Depiction of the difference between fully developed and

developing flow within dialysis catheters. At the arterial inlet port,

blood enters as developing flow rather than parabolic flow, and

this flow has high resistance. The higher the flow rate, the greater

the proportion of developing flow in the catheter lumen (24).

Fig. 10. Effect of side-hole tip geometry on the relationship

between flow and pressure. The more complicated the tip design,

the less developed is the flow pattern and the greater the catheter

resistance (24).

Fig. 11. Clots adherent to the side holes of Mahurkar-type step-

tip catheters, in the study by Tal et al. (25).

ADVANCES IN TUNNELED CVC FOR DIALYSIS 7

catheters had no flow when placed in the SVC of the pigthough the split tip and Palindrome allowed flow.In the patient and over time, any catheter lying on

the vein or atrial wall will be affected by sheaths, clots,and the relation of the catheter tip to the vascular sur-faces, and these factors affect recirculation. In a recentsurvey of recirculation in a dialysis unit, Moossaviet al. measured recirculation in patients with step-tip,split tip and symmetrical (Palindrome) catheters, whilethe catheters were run in the usual flow direction. Alltunneled catheters delivered the same recirculation,between 6% and 8%. Acute dialysis catheters howeverdelivered blood flow with 23% recirculation. Thoughthe Palindrome catheter appears successful, there areno clinical data yet showing that it diminishes recircu-lation or has higher flow rates over time than othercatheters (26).

Self-Centering SVC Catheter, the CentrosTM

The idea of the Centros� catheter is fairly simple.The best way to avoid complete sheathing of the catheteris to hold a portion of the catheter in the middle of thevena cava. As shown in Fig. 2, the Centros� catheterhas a distal end that is planar with two outward bends of

the distal tips that create two points of contact with theSVC. The ports bend inward from the contact points.The contact points center the catheter in the vena cavaand place the ports in a position away from the wall ofthe vein wall. Side holes are not necessary as the portsshould not be in contact with the vena cava wall andtherefore should not become occluded. If fibrous sheath-ing near the ports is avoided the blood flow rate of theCentros� should be maintained over many months ofuse, and recirculation should be minimal (in vitro testsdemonstrate zero recirculation). The catheter is designedto be placed in the lower third of the SVC rather thanwithin the atrium, so placement is made easier. In someversions a gently preformed curve of the apex of thecathetermatches the usual arc of the subcutaneous tract.There is some prior evidence that a catheter sup-

ported within the SVC will remain free of sheathingand thrombosis. In 1998 Kohler and Kirkman reportedan animal trial in which single lumen 3.2 mm diametercatheters were placed in the SVC of pigs, and left for 1–8 weeks (27). No anticoagulant was administered andthe catheters were not used for infusion or bloodremoval. Some of the catheters had a 2 cm diameterloop attached to them, to center the tip in the distalSVC. During placement of the catheters without a loop,fluoroscopy demonstrated a continued relative motionof the catheter and vena cava wall with each heartbeat;however, catheters with the loop remained stationary atthe point of contact with the vena cava wall. As shownin Fig. 14 when the loop catheters were examined at theend of the 8 week period, the SVC and catheter werecompletely free of fibrous sheathing and thrombosis. Inpigs with nonlooped catheters the catheter was com-pletely covered by sheath and thrombus, the SVC wasnearly occluded, and there was a much greater numberand size of intimal lesions. From this study it is appar-ent that a catheter which is supported in the vena cavaby two points of contact should have considerably lesssheathing and fibrosis, and therefore more constantflow over time, versus straight, single-body or split-tipcatheters.In June 2007 the FDA approved the Centros� cathe-

ter for use in dialysis access. A small number of these

Fig. 12. The symmetrical Palindrome catheter of Dr. Tal, with bias-cut ports. Kinetic energy carries returning blood downstream, and

pressure gradient brings blood through the closest part of the removal blood lumen (14).

Fig. 13. Recirculation rates with reversed flow for step tip, split

tip and Palindrome catheters, in an animal model (14).

8 Ash

catheters was created of 28 and 24 cm length (hub to tip)and a gentle bend at the expected apex (as in Fig. 2). Allcatheters had a self-sealing opening on the venous limbto allow threading over a single guidewire, though thisfeature is not needed for placement through a splitsheath. After IRB approval we initiated a clinical trial atfour independent centers (Ash SR, Clarian ArnettHealth, Lafayette, IN; Asif A, University of Miami,Miami, FL; Yevzlin AS,University ofWisconsin,Madi-son,WI; SamahaAL,Kidney andHypertensionCenter,Cincinnati, OH) to determine the hydraulic properties ofthe catheter and permanence of flow. Inclusion criteriawere quite broad:

• Adult patients initiating or continuing in-centerHD who were scheduled to receive a tunneleddialysis catheter in the right IJ vein;

• Patients who may have had a prior tunneledcatheter for dialysis in the right IJ site wereeligible for inclusion in the study; and

• Patients were expected to require use of the cath-eter for more than 45 days.

Exclusion criteria were few:• Patients with a prior history of right IJ vein

thrombosis, or• With documented stenosis of the right IJ,

innominate, or SVC;• Patients in whom the catheter would be placed

into the same site as an existing catheter; and• Patients who were unable to sign a consent

form.Nine catheters were placed into patients, in all cases

through the right IJ, under local anesthesia using fluo-roscopy, ultrasound, and a 16 French split sheath. Tipsof the catheters were placed in the lower third of theSVC rather than in the atrium. An example of a post-placement X-ray is in Fig. 15. The catheters were usedthree times per week for outpatient hemodialysis treat-ments. Once per week at the start of dialysis, the bloodpump was set to deliver a negative arterial pressure of)200 mmHg on the arterial line. The blood flow rateassociated with this modestly negative pressure wasrecorded (Qb)200). The same measurement was also per-formed on 120 prevalent DD tunneled CVC for dialysisin 20 dialysis centers, as part of a study of catheter locks.In the current study, reasons for catheter removal wererecorded as were any significant problems with thecatheters.

Figure 16 demonstrates the average Qb)200 flow ratefor all nine catheters, over the 7-week study. The meanQb)200 was 390 ml ⁄minute (SD � 49) at catheter inser-tion and was 401 ml ⁄minute (SD � 80) at 7 weeks ofuse (NS). By comparison, 120 standard tunneled dialysisIJ catheters being used in several dialysis units had alower Qb)200, 348 ml ⁄minute (SD � 64, p < 0.05).During 7 weeks of follow-up, one of the nine self-center-ing catheters was removed due to presumed exit

(a)

(b)

(c)

Fig. 14. (a) Catheter design of Dr. Kohler with loop to support

the distal tip in the middle of the vena cava. (b) Position of loop in

SVC. (c) Appearance of SVC near tip of catheter, 8 weeks after

placement (27).

Fig. 15. X-ray of chest showing placement of the Centros tips in

the lower third of the SVC.

ADVANCES IN TUNNELED CVC FOR DIALYSIS 9

infection, and one was removed when no longer needed.No catheter failed to provide adequate flow for dialysisduring the study.At the end of the study catheters continued to be

used for dialysis, in some patients for up to 6 months.Two patients volunteered for a computed tomography(CT) scan of the chest after 4 months of catheter use.Figure 17 demonstrates that the catheters were in theexpected position, limbs in a plane across the venacava with at least one distal limb separated from the

wall of the vena cava. No sheath was seen but ofcourse a standard CT could not detect sheaths of lessthan 1 mm thickness. For comparison, Fig. 18includes two CT scans performed on patients withSplit Cath catheters, demonstrating that both limbs ofthese catheters are adjacent to the vena cava wall, as isalways the case. One of these Split Cath catheters hasa clot or sheath covering the catheter. At removal,none of the Centros� catheters exhibited any resis-tance to retraction, and none came out with remnants

Fig. 16. Flow rate of the Centros catheters at modestly negative arterial side pressure (QB)200), over 7 weeks of dialysis use. For compari-

son, same measurement carried out on 120 prevalent DD tunneled dialysis catheters.

Fig. 17. CT scans of the chest taken 4 months after Centros placement in two patients, demonstrating that the catheter limbs form a

plane within the vena cava and at least one port is separated from the vena cava wall. The SVC on the right includes two pacemaker wires.

Fig. 18. CT scans of two patients with Split Cath catheters in place for several months. Note that both limbs of the catheter lie against

the vena cava wall. Clot surrounds the catheter in the CT angiogram on the right.

10 Ash

of any thickened sheath. In fact most catheters wereremarkably free of any clots or evidence of sheath asshown in Fig. 19.

This preliminary study indicates that the self-centeringCentros� catheter provides highly acceptable flow rateat modest negative pressure, without deterioration inflow rate over 7 weeks of use. This high flow rateoccurred despite positioning of the tips of the catheter inthe SVC (rather than within the atrium). The Centros�catheter is now being marketed widely. The studyshowed no evidence of any significant sheathing of thecatheters. Two large multi-center studies are planned forthe catheter, one observational and one randomizedversus a standard split-tip catheter.

Current Challenges and Future Directions forTunneled CVC for Dialysis

The advent of successful tunneled CVC for dialysishas been a great advance for patients with ESRD,both for beginning hemodialysis and for continuingdialysis. Tunneled CVC now allow dialytic supportof patients for many months if needed, allowingpatients to be supported long enough for fistulas and

grafts to be created, corrected, and become the bestlong-term access choices.

In spite of advances, tunneled CVC still have signifi-cant problems and limitations. For each of these prob-lems, there will someday exist a solution which willadvance the technology and benefits of tunneled CVCfor dialysis:

• Catheter-related infections: Catheter materials,chemical impregnation methods, or catheterlocks are now being investigated to kill bacteriain the biofilm layers both on the outside andinside of tunneled CVC, in order to decrease thismost common complication of the catheters.Two recent reviews have confirmed that everyantibacterial catheter lock that has been studiedin randomized, prospectively controlled trialshas demonstrated a 50–80% decrease in the inci-dence of CRBSI (28,29). The antibacterial effectof impregnated chemicals and new cathetermaterials in a tunneled CVC must remain formany months rather than a week or so (as withacute catheters), and this implies the need forsome method of regeneration of the active com-ponent over time, or use of covalently boundenzymatic or catalytic materials.

• Catheter tip clotting: As mentioned above, cath-eters without side holes have some advantage inavoiding clotting. Catheters which open andclose at the tip would allow the catheter to retainanticoagulant and completely avoid blood clot-ting within the ports.

• Catheter fibrous sheathing: As described above,one solution for catheter sheathing may be acatheter which centers itself in the vena cava(the Centros�). Other approaches includechemical impregnation of the catheter to preventthe growth of macrophages and fibroblastsaround the catheter bodies, though this is likelyto be difficult as the irritation of the vein wall bythe catheter is such a strong stimulus for sheathformation.

• Central venous stenosis: Methods to distribute ordiminish ‘‘wear’’ on the vena cava must beevolved to avoid this serious and still frequentcomplication. Avoiding use of acute dialysiscatheters diminishes the frequency of centralvenous stenosis. Catheters that are supported atonly two points in the SVC diminish the contactarea of the catheter to the vein but mightincrease wear at two points. Whether these cath-eters diminish the risk of SVC stenosis in thelong run is unclear.

• External component bulk: Patients bandage andkeep dry the hubs, extension tubings, clamps,and connectors, but many also complain aboutthe general bulk of the catheter components ontheir bodies. Also, the preclusion of showering isa real bother to many patients. Subcutaneousports were proposed as one solution (LifeSite,BioLink) but clearly are not the answer for mostlong-term patients. Eventually more radicalskin-level ‘‘connectology’’ will be necessary.

Fig. 19. Appearance of Centros� catheters after removal and

gently rinsing with sterile saline. The only signs of any thin

sheath was in a few catheters, near the jugular vein entrance

sheath.

ADVANCES IN TUNNELED CVC FOR DIALYSIS 11

• External component breakage: More durable yetstill light-weight components are possible. Sim-plifying the entire catheter design to limit thesize and number of glued connections is a partialsolution.

With a few more improvements, tunneled CVC fordialysis could become a painless, effective and safelong-term access for the majority of dialysis patientsand perfectly acceptable as an alternative to AV grafts.For those patients in whom they are possible, the fistulawill likely remain the optimal access for some years.

Conflict of Interest

Dr. Ash is Chairman and Director of Research forAsh Access Technology, Inc., the originator of the SplitCath� andCentros� catheter designs.

References

1. Beathard GA: Strategy for maximizing the use of arteriovenous fistulae.Semin Dial 13:291–296, 2000

2. 2006 Annual Report: ESRD Clinical Performance Measures Project.Department of Health and Human Services, Centers for Medicare &Medicaid Services, Office of Clinical Standards & Quality, Baltimore,MD: CMS, 2006

3. Beathard GA: ASDIN Curriculum. Clinton, MS, ASDIN, 20084. Beathard GA: KDOQI Guideline updates and clinical practice guide-

lines 2006. CPG 2.4 (Basic Principles for Catheter Insertions), NewYork, National Kidney Foundation

5. Ash SR: Fluid mechanics and clinical success of central venous cathetersfor dialysis – answers to simple but persisting problems. Semin Dial20:237–256, 2007

6. Schwab SJ, Beathard G: The hemodialysis conundrum: hate living withthem, but can’t live without them. Kidney Int 56:1–17, 1999

7. Foraner AR, Theohasis C: Histologic changes in the human vein walladjacent to indwelling central venous catheters. J Vasc Interv Radiol14:1163–1168, 2003

8. D’Othee BJ, Than JC, Sheiman RG: Restoration of patency in failinghemodialysis catheters. J Vasc Interv Radiol 17:1011–1015, 2006

9. Canaud B, Leray-Moragues H, Garrigues V, Mion C: Permanent twincatheter: a vascular access option of choice for haemodialysis in elderlypatients. Nephrol Dial Transplant 7:82–88, 1998

10. Canaud B, Leray-Moragues H, Kamoun K, Garrigue V: Temporaryvascular access for extracorporeal therapies. Ther Apher 4:249–255,2000

11. Blake PG, Huraib S, Wu G, Uldall PR: The use of dual lumen jugularvenous catheters as definitive long term access for haemodialysis. Int JArtif Organs 13:26–31, 1990

12. Schwab SJ, Buller GL, McCann RL, Bollinger RR, Stickel DL:Prospective evaluation of a Dacron cuffed hemodialysis catheter forprolonged use. Am J Kidney Dis 1:166–169, 1988

13. Tapson JS, Hoenich NA, Wilkinson R, Ward MK: Dual lumensubclavian catheters for haemodialysis. Int J Artif Organs 8:195–200,1985

14. Tal MG: Comparison of recirculation percentage of the palindromecatheter and standard hemodialysis catheters in a swine model. J VascInterv Radiol 16:1237–1240, 2005

15. Ash SR, Brown K, Asif A, Yevzlin S, Samaha AL: Preliminary clinicalstudy of a self-centering central venous catheter for dialysis (Centros�).American Society of Diagnostic and Interventional Nephrology(ASDIN) Accepted Abstracts – Fourth Annual ASDIN Scientific Meet-ing February 9–10, 2008. Semin Dial 21:105–111, 2008

16. Mareels G, Kaminsky R, Verdonck PR: Particle image velocimetry-vali-dated, computational fluid dynamics-based design to reduce shear stressand residence time in central venous hemodialysis catheters. ASAIO J53:438–446, 2007

17. Trerotola SO, Shah H, Johnson M, Namyslowski J, Moresco K, PatelN, Kraus M, Gassensmith C, Ambrosius WT: Randomized comparisonof high-flow versus conventional hemodialysis catheters. J Vasc IntervRadiol 10:1032–1038, 1999

18. Richard HM, Hastings GS, Boyd-Kranis RL, Murthy R, Radack DM,Santilli JG, Ostergaard C, Coldwell DM: A randomized, prospectiveevaluation of the Tesio, Ash Split, and Opti-flow hemodialysis catheters.J Vasc Interv Radiol 12:431–435, 2001

19. Trerotola SO, Kraus M, Shah H, Namyslowski J, Johnson MS, SteckerMS, Ahmad I, McLennan G, Patel NH, O’Brien E, Lane KA, Ambro-sius WT: Randomized comparison of split tip versus step tip high flowhemodialysis catheters. Kidney Int 62:282–289, 2002

20. Wang J, LaBerge JM, Chertow GM, Kerlan RK: Tesio catheter accessfor long-term maintenance hemodialysis. Radiology 241:284–290, 2006

21. Polaschegg HD, Shah C: Overspill of catheter locking solution: safetyand efficacy aspects. ASAIO J 49:713–715, 2003

22. Polaschegg HD: Loss of catheter locking solution caused by fluiddensity. ASAIO J 51:230, 2005

23. Kindgen-Milles D, Kram R, Kleinekofort W: Assessment of temporarydialysis catheter performance on the basis of flow and pressure measure-ments in vivo and in vitro. ASAIO J 53:351–356, 2007

24. Fricker ZP, Rockwell DO: Pressure drop through generic lumens ofhemodialysis catheters. ASAIO J 53:428–433, 2007

25. Tal MG, Peixoto AJ, Crowley ST, Denbow N, Eliseo D, Pollak J: Com-parison of side hole versus non side hole high flow hemodialysis cathe-ters. Hemodial Int 10:63–67, 2006

26. Moossavi S, Kaufman T, Jordan J, Moossavi S, Vachharajani T: Retro-spective evaluation of factors influencing the catheter recirculation inprevalent dialysis patients. Semin Dial 21:197–199, 2008

27. Kohler TR, Kirkman TR: Central venous catheter failure is induced byinjury and can be prevented by stabilizing the catheter tip. J Vasc Surg28:59–65, 1998

28. Jaffer Y, Selby NM, Taal MW, Fluck RJ, Mclntyre CW: A meta-analy-sis of hemodialysis catheter locking solutions in the prevention of cathe-ter-related infection. Am J Kidney Dis 51:233–241, 2008

29. Labriola L, Crott R, Jadoul M: Preventing haemodialysis catheter-related bacteraemia with an antimicrobial lock solution: a meta-analysisof prospective randomized trials. Nephrol Dial Transplant 23:1666–1672, 2008

12 Ash